lyosha commited on
Commit
54330de
·
verified ·
1 Parent(s): 2634140

Upload VoicePostprocessModel DACVAE checkpoints

Browse files
VoicePostprocessModel/README.md ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # VoicePostprocessModel Checkpoints
2
+
3
+ Trained checkpoints for the DACVAE/PCA recipe in VoicePostprocessModel.
4
+
5
+ Files:
6
+ - `dacvae_estimator_best.pt`: differentiable acoustic-state estimator over Echo-TTS Fish/DAC + PCA latents.
7
+ - `dacvae_editor_general_best.pt`: broad dry-recovery editor trained on room, EQ, distortion, compression, bandwidth, and noise variants.
8
+ - `dacvae_editor_target_reverb_best.pt`: target-only editor trained on the calibrated room/bathroom reverb suite.
9
+ - `dacvae_editor_mixed_corrective_best.pt`: mixed corrective editor fine-tuned on the calibrated suite plus the broad cache.
10
+
11
+ Latent space dependencies:
12
+ - Fish/DAC checkpoint: https://huggingface.co/jordand/fish-s1-dac-min
13
+ - Echo-TTS PCA state: https://huggingface.co/jordand/echo-tts-base/blob/main/pca_state.safetensors
VoicePostprocessModel/dacvae_editor_general_best.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e687a8fd7558a2e1888205bb9bb27e4de967475094a1b176b4ce222b6dc01bf9
3
+ size 21545193
VoicePostprocessModel/dacvae_editor_mixed_corrective_best.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f0d7fd9a2a5807908e598dfc3a6d4b4911e3d775882d8bb585663483e3c6912
3
+ size 21545193
VoicePostprocessModel/dacvae_editor_target_reverb_best.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:54a91e5a4e2e193422a60c278eaff821b2755268ee70acf6be9cd259bb28bd9e
3
+ size 21545193
VoicePostprocessModel/manifest.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "project": "VoicePostprocessModel",
3
+ "latent_space": "Echo-TTS Fish/DAC z_q projected through Echo-TTS PCA to 80 dimensions",
4
+ "sample_rate": 44100,
5
+ "files": {
6
+ "estimator": "dacvae_estimator_best.pt",
7
+ "editor_general": "dacvae_editor_general_best.pt",
8
+ "editor_target_reverb": "dacvae_editor_target_reverb_best.pt",
9
+ "editor_mixed_corrective": "dacvae_editor_mixed_corrective_best.pt"
10
+ },
11
+ "external_latent_space_assets": {
12
+ "fish_dac": "https://huggingface.co/jordand/fish-s1-dac-min",
13
+ "echo_tts_pca_state": "https://huggingface.co/jordand/echo-tts-base/blob/main/pca_state.safetensors",
14
+ "echo_tts_repo": "https://github.com/jordandarefsky/echo-tts"
15
+ }
16
+ }